Fit quality

SVI arbitrage constraints.

The dashboard is labelled arbitrage-aware because the upstream pricing engine publishes fitted SVI curves together with diagnostics for no-arbitrage conditions.

The browser displays those fit results and g-test style checks for the live BTC surface. Other maintained currency configurations remain testing scope.

Butterfly checks / Calendar checks / G-test / Fit diagnostics.

By Sean GordonPublished Updated Arb ConstraintsSurface cluster

Purpose

What the constraints protect against.

An options surface can look smooth while still implying impossible prices. The fit layer checks the SVI smile and the surface assembled across expiries before the UI treats it as a production-quality curve.

Smile checks

Butterfly arbitrage checks.

For each expiry, the fitted total variance smile should not imply a negative risk-neutral density. In practical terms, the engine evaluates the smile shape across a grid of log-moneyness points and publishes g-test diagnostics when local convexity or wing behavior becomes suspicious.

Term checks

Calendar arbitrage checks.

At the same moneyness, total variance should generally not fall as expiry increases. The surface grid and tenor rows make that check visible by comparing the fitted BTC smiles across maturities rather than only one expiry at a time.

Dashboard

What the Fit page shows.

The Fit tab exposes the fitted variance curve, fitted volatility curve, 3D surface projection, and g-test panel. If the g-test view spikes or the fit error deteriorates, it is a signal to inspect the market data, stale venue overlays, or calibration inputs before relying on the surface.

SVI-through matrix highlighting bid and ask levels that sit through the fitted implied volatility mid
The SVI-through matrix is a practical market-data check: highlighted cells show where quoted bid or ask levels sit through the fitted surface.
Derivasys fit diagnostics and risk analytics dashboard for SVI volatility surface monitoring
Fit diagnostics are useful only when read alongside live risk analytics, quote quality, and feed state.

Native object

Variance lines and g-test diagnostics.

The variance view plots total implied variance by expiry rather than annualized implied volatility. SVI is calibrated in variance space, so these lines are the native fitted objects.

The g-test view is a diagnostic series published by the upstream engine for the fitted SVI smile. A stable line is expected for a usable fit; spikes or sign changes should be reviewed alongside stale quotes, crossed markets, and current fit error.

Access

API outputs for testing.

The same fit-quality state is available in request-based WebSocket evaluation for the live BTC surface. REST behavior and other configured currencies remain testing scope and are not presented as stable public coverage.

Inspect live surface state in Derivasys.

Use the dashboard to compare fitted surfaces, venue marks, risk nodes, quote diagnostics, and API-ready state.